THE APPLICATION OF VISUAL-BASED ACUPOINT POSITIONING TECHNOLOGY IN ACUPOINT ROBOTS
Keywords:
Acupuncture robot, Acupoint localization, YOLOv8, Trajectory planning, Machine visionAbstract
Acupuncture efficacy depends on the precision of acupoint localization. Based on prior research, this paper proposes a technical framework for automatic facial acupoint positioning and robotic acupuncture: facial information is captured using RGB-D vision, and an improved YOLOv8 model detects 27 index acupoints across 15 categories, enabling precise three-dimensional coordinate mapping for localization. A specialized end-effector integrating needle exchange, holding, and rotation functions converts lifting, thrusting, twisting, retention, and withdrawal into linear and rotational motions. An enhanced APF-RRT* algorithm is proposed to achieve obstacle-avoidance path planning during continuous multi-acupoint needling. Experiments conducted with UR5e, Kinect V2, and a simulated human head demonstrate that the system achieves average errors of 0.54 mm, 0.50 mm, and 0.66 mm in the three dimensions, achieving millimeter-level accuracy in acupoint localization.References
[1] World Health Organization. Acupuncture: review and analysis of reports on controlled clinical trials. Geneva: World Health Organization, 2002: 1-87.
[2] Vickers AJ, Vertosick EA, Lewith G, et al. Acupuncture for chronic pain: update of an individual patient data meta-analysis. J. Pain, 2018, 19(5): 455-474.
[3] Lan KC, Lee CY, Lee GS, et al. An initial study on automated acupoint positioning for laser acupuncture. Evid.-Based Complement. Altern. Med., 2022: 8997051.
[4] Zhang M, Schulze J P, Zhang D. FaceAtlasAR: atlas of facial acupuncture points in augmented reality. Comput. Sci. Inf. Technol., 2021, 11(7): 1-11.
[5] Yuan Z, Shao P, Li J, et al. YOLOv8-ACU: improved YOLOv8-pose for facial acupoint detection. Front. Neurorobot., 2024, 18: 1355857.
[6] Malekroodi HS, Seo SD, Choi J, et al. Real-time location of acupuncture points based on anatomical landmarks and pose estimation models. Front. Neurorobot., 2024, 18: 1484038.
[7] Wang H, Liu L, Wang Y, et al. Hand acupuncture point localization method based on a dual-attention mechanism and cascade network model. Biomed. Opt. Express, 2023, 14(11): 5965-5978.
[8] Liu YB, Qin JH, Zeng GF. Back acupoint location method based on prior information and deep learning. Int. J. Numer. Method Biomed. Eng., 2023, 39(12): e3776.
[9] Karaman S, Frazzoli E. Sampling-based algorithms for optimal motion planning. Int. J. Robot. Res., 2011, 30(7): 846-894.
[10] Gammell JD, Srinivasa SS, Barfoot TD. Informed RRT*: optimal sampling-based path planning focused via direct sampling of an admissible ellipsoidal heuristic. Proc. IEEE/RSJ Int. Conf. Intell. Robots Syst., 2014: 2997-3004.
[11] Khatib O. Real-time obstacle avoidance for manipulators and mobile robots. Int. J. Robot. Res., 1986, 5(1): 90-98.
[12] Terven J, Cordova-Esparza DM, Romero-Gonzalez JA. A comprehensive review of YOLO architectures in computer vision: from YOLOv1 to YOLOv8 and YOLO-NAS. Mach. Learn. Knowl. Extr., 2023, 5(4): 1680-1716.
[13] Han K, Wang Y, Tian Q, et al. GhostNet: more features from cheap operations. Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit., 2020: 1580-1589.
[14] Pagliari D, Pinto L. Calibration of Kinect for Xbox One and comparison between the two generations of Microsoft sensors. Sensors, 2015, 15(11): 27569-27589.
[15] Universal Robots A/S. UR5e e-Series datasheet. Odense: Universal Robots A/S, 2024.
[16] Zhang Z. A flexible new technique for camera calibration. IEEE Trans. Pattern Anal. Mach. Intell., 2000, 22(11): 1330-1334.
[17] Tsai RY, Lenz RK. A new technique for fully autonomous and efficient 3D robotics hand/eye calibration. IEEE Trans. Robot. Autom., 1989, 5(3): 345-358.